A satellite blind spot masks crop residue burning across northern India

Ivar Roderick van der Velde , Piyushkumar N. Patel, Ritesh Gautam, Guido R. van der Werf, Gareth Roberts, Ilse Aben 

Postmonsoon crop residue burning in the Indian states of Punjab and Haryana is a major source of air pollution and air quality degradation across the Indo-Gangetic Plain (IGP). Recent satellite observations have revealed an apparent inconsistency between declining fire detections and persistent or increasing atmospheric pollution. Here we combine eight years (2018-2025) of fire observations from low Earth orbiting (LEO) and geostationary satellites, satellite-derived atmospheric composition observations, ground-based air quality observations, meteorological reanalyses data, and chemical transport model simulations to investigate the drivers of this discrepancy. MODIS and VIIRS LEO observations indicate more than 90% decline in fire activity after 2021, whereas the geostationary SEVIRI observations reveal sustained or increasing fire activity. This divergence results primarily from fires started later in the day, beyond the time window the MODIS and VIIRS satellites pass over. Moreover, despite extensive burning, atmospheric pollution levels varied substantially between years. In Punjab, carbon monoxide and aerosol concentrations reached record levels in 2024 but were comparatively modest in 2025. Meteorological analyses indicate that persistent eastward winds in 2025 enhanced the transport of clean air into the IGP, substantially reducing atmospheric accumulation of pollutants. We conclude that changes in burn timing can substantially bias conventional fire monitoring systems and that meteorological variability and observational constraints imposed by cloud cover can be as important as emissions in interpreting regional air quality impacts.